CALL FOR PAPERS – Map and spatial pattern comparison
Annual Meeting of the Association of American Geographers, San Francisco, 2016
Organizers: Colin Robertson (Wilfrid Laurier), Jed Long (St. Andrews)
Spatial patterns have gone mainstream. Every field from urban planning to archeology to history are collecting and utilizing spatial information. Corporations collect vast amounts of data tied to different geographical scales that are exploited for profit. One of the core questions driving many new applications of spatial data is – when are two spatial patterns similar?
This deceptively simple question hides complexity that goes to the root of core GIScience and spatial analysis methodology. What does it mean for spatial patterns to be similar? The question of spatial pattern similarity has most thoroughly been investigated for landcover maps. Frameworks have been introduced for comparing categorical maps as the outcomes of spatial processes (Csillag and Boots 2005) and for geosimulation models (Hagen-Zanker and Martens 2011) – developing algorithms and significance tests that can explain whether differences between two maps are due to chance, or could have been generated by the same process. Spatial pattern similarity has also been explored in the context of landscape patterns through the comparison of the distribution of pattern indices and their links to spatial process models (Remmel and Csillag 2003). Importantly, statistically distinguishing spatial pattern differences is non-trivial, even for mapped patterns that appear different to a human observer.
Techniques for comparing non-categorical and/or non-lattice forms of spatial data such as continuous raster maps, tessellations, line patterns, point patterns, have been treated in piecemeal and fragmented fashion. For example, movement trajectories have been the subject of numerous methodological innovations in recent years due to the advance of GPS tracking data systems; yet these developments have not been integrated or adapted to other spatial representations. A framework for comparison of spatial patterns in general is currently lacking.
In this session, we will investigate questions pertaining to all aspects of spatial pattern comparison and invite submissions from researchers working on any aspect of spatial pattern/map comparison. Research into the theoretical, applied, and philosophical dimensions of map comparison are all welcome. We are interested in putting together a diverse set of papers that reflect the wide array of approaches being taken in spatial pattern comparison research, including:
Stochastic spatial modelling
Image compression and computer vision
Landscape pattern indices
Movement trajectory analysis
Point pattern analysis and modelling
Polygon change analysis and modelling
Applied pattern comparison
Papers Visualizing spatial patterns Abstracts of 250 words should be should be submitted to both organizers (firstname.lastname@example.org and email@example.com) for consideration for inclusion in the session by October 16th 2015. Participants will be notified by October 23rd and will then need to register for the conference by October 29th in order to be included in the session. Please include the phrase “AAG 2016” in the subject line of your e-mail.